Hepatocyte growth factor (hereinafter abbreviated as HGF) is a protein having a mitogenic activity on hepatocytes. Some differences in amino acid sequences are observed among known HGFs and HGF is also named as SF (scatter factor), TCF (tumor cytotoxic factor) and the like in addition to HGF. The known proteins having mitogenic activities on hepatocytes are collectively named as HGFs in the present invention. HGFs are known to be physiologically active peptides that exert various pharmacological actions such as mitogenic action, morphogenetic action, neovascularization action, nerve protective action and anti-apoptotic action, in addition to mitogenic activity on hepatocytes (see non-patent document 1: Matsumoto, K. et al., Kidney International, 2001, vol. 59, p 2023-2038).
From its pharmacological actions, HGF is expected to be developed as therapeutic agents for cirrhosis, therapeutic agents for renal diseases, epithelial cell proliferation promoters, anti-cancer agents, preventive agents for side effects in cancer therapy, therapeutic agents for lung injury, therapeutic agents for gastroduodenal injuries, therapeutic agents for cerebral neuropathy, preventive agents for immunosuppression side effects, collagen degradation promoters, therapeutic agents for cartilage injury, therapeutic agents for artery diseases, therapeutic agents for pulmonary fibrosis, therapeutic agents for hepatic diseases, therapeutic agents for blood coagulation malfunction, therapeutic agents for plasma hypoproteinemia, therapeutic agent for wounds, neuropathy improving agents, hematopoietic stem cell increasing agents and hair restoration promoters (for example, see patent documents 1 to 14: JP-A-4-18028, JP-A-4-49246, EP-492614-A, JP-A-6-25010, WO93/8821, JP-A-6-172207, JP-A-7-89869, JP-A-6-40934, WO94/2165, JP-A-6-40935, JP-A-6-56692, JP-A-7-41429, WO93/3061 and JP-A-5-213721).
HGF is secreted from organs such as liver, brain, lung, bone marrow, spleen, placenta and kidney, or from blood cells such as platelets and leukocytes. However, since HGF is present in the body in a minute quantity, it is necessary to produce HGF in a large scale using cells by genetic engineering techniques in order to use it as medical preparations. It has been known that HGF can be produced using animal cells such as Chinese hamster ovary (CHO) cells (see, for example, patent documents 15 and 16: JP-11-4696-A and JP-10-191991-A).
However, the method for producing proteins using animal cells such as CHO cells is expensive, resulting in increase of drug prices.
As a method for producing a recombinant protein at a low cost, expression of proteins in prokaryotic cells such as E. coli by introducing genes of interest into them has been known (see non-patent document 2: Swarts, J. R., Current Opinion in Biotechnology, 2001, vol. 12, p 195-201). However, there exists a problem that no glycosylation occurs in recombinant proteins produced in the prokaryotic cells such as E. coli. This is because the prokaryotic cells such as E. coli do not contain endoplasmic reticulum and Golgi apparatus that are places for biosynthesis of sugar chain(s).
Addition of a sugar chain to a protein and its modification in an animal cell is post-translational modifications using no template, differing from the case of biosynthesis of DNAs or proteins. This post-translational modification is performed by a complicated mechanism mediated by various glycosylation-related enzymes locally present in intracellular organelle called endoplasmic reticulum and Golgi apparatus. That is, a sugar chain is elongated so as to obtain a given sugar chain structure when sequential addition and cleavage of monosaccharides occur according to a complicated biosynthetic pathway catalyzed by enzymes specific to certain linkages of monosaccharides (glycosidase and glycosyltransferase) (see non-patent document 3: Kornfeld, R., et. al, Annual Review of Biochemistry, 1985, vol. 54, p 631-664).
Sugar chain(s) added to proteins in this way have been known to be deeply involved in whole life phenomena of higher organisms (see non-patent documents 4 and 5: Kobata, A., European Journal of Biochemistry, 1992, vol. 209, p 483-501; Varki, A., Glycobiology, 1993, vol. 3, p 97-130).
It has been known that half or more of proteins in the human body exist as glycoproteins to which sugar chains are added (see non-patent document 6: Goochee, C. F. et al., Biotechnology, vol. 9, p 1347-1355).
If a glycoprotein originally present in the form carrying sugar chains is converted into a form containing no sugar chain, there is a fear of losing activity. For example, it is known that erythropoietin, known as a hematopoietic hormone, lose its activity when the sugar chains are removed (see non-patent document 7: Takeuchi, M. et al., Glycobiology, 1991, vol. 1, p 337-346).
Yeast is known as a host cell that is capable of producing a recombinant protein at low cost and has a glycosylation ability (see non-patent documents 8: Wiseman, A., Endeavor, 1996, vol. 20, p 130-132; non-patent document 9: Russell, C. et al., Australian Journal of Biotechnology, 1991, vol. 5, p 48-55; non-patent document 10: Buckholz, R. G. et al, Biotechnology, 1991, vol. 9, p 1067-1072). Since yeast is a eukaryotic cell and has endoplasmic reticulum and Golgi apparatus, it is consequently equipped with glycosylation mechanism. However, since the glycosylation mechanism of yeast differs significantly from that of animal cells, when a protein having glycosylation site(s) is produced in yeast, sugar chain(s) of yeast type would be added. It is known that the sugar chain structures of yeast are significantly different from those of human and other mammals (see non-patent document 11: Germmill, T. R. et al., Biochemica et Biophysica Acta, 1999, vol. 1426, p 227-237).
Accordingly, such recombinant proteins cannot be used for medicines for human beings and other mammals because they exhibit antigenicity against human and mammals.
Further, an insect cell is also a host having a glycosylation ability and can produce a protein at relatively low cost, however, the sugar chain structures of an insect cell are also different from those of human type (see non-patent document 12: Altmann, F. et al., Glycocomjugate Journal, 1999, vol. 16, p 109-123).
Accordingly, there is a possibility for a recombinant protein derived from insect cells to show antigenicity against human and other mammals.
Then, one can envisage production of a protein containing no sugar chains by removing sugar chains from a protein produced using yeast, insect cells, or the like, or by introducing a gene designed to have mutation(s) at glycosylation sites in a protein molecule into yeast, insect cells, or the like. However, if a protein originally present in the form carrying sugar chains is converted into a protein containing no sugar chain, there is a fear of losing activity, as described above.
Five sugar chains are added to HGF (see non-patent documents 13: Hara, H. et al., Journal of Biochemistry, 1993, vol. 114, p 76-82; non-patent documents 14: Shimizu, N, et. al, Biochemical and Biophysical Research Communications, 1992, vol. 189, p 1329-1335). With respect to the influence of removing the sugar chains of HGF on the activity, it is reported that, when HGF-producing cells were cultured in the presence of tunicamycin, an inhibitor of N-glycosylation, secreted HGF maintained the motogenic activity (see non-patent document 15; Hofmann, R. et al., Biochemica et Biophysica Acta, 1992, vol. 1120, p 343-350. HFG is denoted as SF in the report).
However, this report does not give sufficient information since the extent of deficiency of the sugar chains in the HGF produced in the presence of tunicamycin was not analysed.
The report described that HGF maintained motogenic activity after treatment of the HGF with N-glycanase or O-glycanase, however, the report showed that HGF treated with N-glycanase or O-glycanase adsorbed onto a ConA column that recognizes sugar chains. The fact that HGF treated with N-glycanase or O-glycanase adsorbed onto a ConA column means that the removal of the sugar chains was limited. Therefore, the descriptions that the HGF treated with N-glycanase or O-glycanase maintained motogenic activity does not lead to a conclusion that glycosylation-deficient HGF maintains motogenic activity.
HGF has a variety of activities, including mitogenic activity, morphogenetic activity, neovascularization activity, anti-apoptoptic activity and nerve protective activity in addition to the motogenic activity (see non-patent document 16: Matsumoto, K. et al., Biochemical and Biophysical Research Communications, 1997, vol. 239, p 639-644).
It cannot be always concluded that functions of HGF other than motogenic activity are maintained even if glycosylation-deficient HGF retains the motogenic activity. For example, NK2 that is a truncated variant of HGF has motogenic activity, whereas it has no mitogenic activity (see non-patent document 17: Hartmann, G. et al., Proceedings of National Academy of Science of the United States of America, 1992, vol. 89, p 11574-11578).
As can be seen from the above, it was unclear at all how many of the diverse functions are maintained in non-glycosylated HGF. HGF has been considered to be a repair factor of organs because of its diverse activities, and it could not be concluded that highly complicated functions are not affected by deficiency of sugar chains in HGF molecules.